Autism Biomedical Information Network

Unproven treatments

Treatments advocated for autism in which data regarding efficacy have not been replicated in peer-reviewed medical journals or for which there are conflicting data:

  • Auditory integration training

  • Facilitated communication

  • Hyperbaric oxygen

  • Secretin

  • Vitamin B6 and magnesium

  • Dimethylglycine (DMG)

  • Intravenous immunoglobulin (IVIG)

  • AZT (zidovudine, Retrovir)

  • Steroids

  • Antifungal medications

  • Detoxication; chelation

  • Dietary manipulations (elimination of gluten, casein, etc.)

  • Hippotherapy; dolphin therapy

  • Sensory integration therapy

  • Craniosacral therapy

  • Behavioral optometry


Most of the proponents of unproven therapies base their claims of efficacy on unstructured assessments or subjective, impressionistic, anecdotal reports as to changes in behavior. These include reports using such imprecise terms as "improved eye contact," "improved social interaction," and "improved language." Unfortunately behavioral effects are exceedingly difficult to measure but objective tests by impartial observers are the gold standard for credible data.

The gold standard for assessing the possible benefit of a proposed treatment is a controlled trial designed to eliminate observer bias. The challenge for autism treatment research is that there is no simple test that can be quantified before and after an intervention. It is necessary to have an instrument capable of measuring a possible change in behavior. Treatment effects (or lack of them) can be assessed using behavior rating scales designed specifically for autism and standardized for reliability when administered by trained observers. The characteristics of a well-designed trial for a medical intervention (usually a psychopharmacotherapy intervention) are:

  • Random assignment to a treatment and a control group
  • A test to show that both groups are otherwise comparable in terms of diagnosis, age, gender, history of prior treatments, history of other concurrent interventions, etc.
  • Double-blind
  • Placebo-controlled
  • Cross-over from treatment to placebo and from placebo to treatment group
  • Use of a behavior rating scale validated for serial use
  • Use of a behavior rating scale by a trained observer (example: the ADOS requires formal training)
  • Publication in a peer-review professional journal

An example of a well-designed study of a medical intervention is the multicenter trial of secretin: Medscape General Medicine on the Web. The abstract of the above study is at: Abstract: Multicenter trial of secretin

There are also well-accepted methods for evaluating the effects of an intensive behavioral intervention in an individual subject. An excellent general reference is:

Behavioral intervention for young children with autism. A manual for parents and professionals. Maurice C (ed) and Green G Luce SC (co-ed), PRO-ED, 1996 (8700 Shoal Creek Road, Austin, TX 78757-6897).

See Chapter 2 (Gina Green), "Evaluating claims about treatments for autism" and Chapter 4 (Tristram Smith), "Are other treatments effective?" (includes discussion of "biological treatments")

What about "biological treatments"?
It is often implied that there is an inherent superiority to so-called "biological treatments" compared to the treatments of "traditional medicine." There is no scientific basis for this presumption since many "traditional" medications are derived from "natural" sources. A few examples are:
  • Penicillin, from certain molds
  • Digoxin, related to compounds extracted from the foxglove plant
  • Salicylate (methyl salicylate from oil of wintergreen)

An excellent resource is the recent (1999) 3-volume publication of the New York State Department of Health Early Intervention Program, Clinical Practice Guideline for Assessment and Intervention for Young Children (Age 0-3 years) with Autism/Pervasive Developmental Disorders (New York State Department of Health Recommendations re Diagnosis and Treatment of Autism ). The evidence-based guideline evolved after a comprehensive review of the literature by an independent panel of professionals and parents. The two interventions which meet an evidence-based standard are:

  1. Early intensive behavioral intervention
  2. Psychopharmacotherapy

It is acknowledged that psychopharmacotherapy does not resolve the core disturbance in autism but can help to ameliorate certain behaviors such as hyperarousal, aggression/oppositional-defiant behavior, perseveration, anxiety, self-induced injury, and other problematic behaviors. Psychopharmacology is on the threshold of directing research in more specific "laser-guided" ways at specific neuronal nets or neurotransmitter receptor subpopulations. This holds out the promise of better treatments that do not result in a more generalized side-effect profile.

These publications can be ordered from:

Health Education Services
P.O. Box 7126
Albany, New York 12224

The price for non-New York State residents is: $21.50

Telephone orders: 518-439-7286


American Academy of Pediatrics, Committee on Children with Disabilities, Policy Statement on Auditory Integration Training and Facilitated Communication for Autism, published in PEDIATRICS 1998;102:431 (August)

Tolbert L et al, Brief report: Lack of response in an autistic population to a low dose clinical trial of pyridoxine plus magnesium. J Autism Dev Disorders 1993;23:193-199.

Pfeiffer SL, Efficacy of vitamin B6 and magnesium in the treatment of autism: A methodology review and summary of outcomes. J Autism Dev Disorders 1995;25:481-493.

Findling RL et al, High-dose pyridoxine and magnesium administration in children with autistic disorder: An absence of salutary effects in a double-blind, placebo-controlled study. J Autism Dev Disorders 1997;27:467-478.

DelGiudice-Asch G et al, Brief report: A pilot open clinical trial of intravenous immunoglobulin in childhood autism. J Autism Dev Disorders 1999;29:157-160.

Bolman WM Richmond JA, A double-blind, placebo-controlled, crossover pilot trial of low dose dimethylglycine in patients with autistic disorder. J Autism Dev Disorders 1999;29:191-194.

|Return to list of commentaries| |Links to other autism resources|

Commentary by Ronald J. Kallen, M.D., ©1999;2000
This page uploaded on 2/11/99; last updated on 6/09/00